> ## Documentation Index
> Fetch the complete documentation index at: https://docs.blockdb.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Free Research Datasets

> Free, audit-grade Ethereum datasets for researchers, quants, and AI builders — 10+ years of token transfers and DEX swaps, published openly with full lineage on Hugging Face.

## Public chain data should be public

Querying Ethereum's history shouldn't require a corporate budget. The data is already on-chain — yet getting it clean, decoded, and usable still means archive nodes, tracing pipelines, or per-megabyte export bills from analytics platforms.

So we did the heavy lifting and gave it away. **10+ years of Ethereum, indexed, decoded, and lineage-tagged** — free to download, no credit card, no sales call, no API quota.

<Card title="Get the datasets on Hugging Face" icon="database" href="https://huggingface.co/collections/BlockDB/blockdb-full-evm-research-datasets-10-years" horizontal>
  BlockDB Full EVM Research Datasets — 10+ years. Browse, preview, and download every published set.
</Card>

Every row carries a `_tracing_id` back to on-chain evidence. You don't have to trust us — you can verify it through the [Lineage API](/api-reference/evm/lineage/overview).

Files are partitioned Parquet, so you can pull a single token, month, or exchange instead of the full archive.

## Load it

```python theme={null}
import pandas as pd

df = pd.read_parquet(
    "hf://datasets/BlockDB/Stablecoin-Transfers-Ethereum-Cryptocurrency-Data",
)
print(df.head())
```

Full mirror of any dataset:

```bash theme={null}
huggingface-cli download BlockDB/Stablecoin-Transfers-Ethereum-Cryptocurrency-Data \
  --repo-type dataset --local-dir ./blockdb-stablecoin-transfers
```

## Licensing

* **Free to use** for research, backtesting, model training, and commercial products built on top of the data.
* **Keep the `_tracing_id` column** if you redistribute — it preserves provenance back to on-chain evidence.

## Need more than this snapshot?

These are historical snapshots. When you need to go live:

* **Filters, other tokens, or recent history** — [BlockDB Historic REST API](https://docs.blockdb.io/api-reference/overview/home).
* **Real-time streams** — [BlockDB WebSocket feed](https://docs.blockdb.io/wss-reference/overview/introduction).
* **Custom extracts** (anything we can derive from on-chain data) — [support@blockdb.io](mailto:support@blockdb.io).